✦ Our pick · UI & web pages
Claude Opus 5.5 at medium effort, $17.55 per finished task, is the cheapest you can get today. Within our measurement error it ties with Claude Sonnet 5.5 and GPT-6.1 Sol: equally good bets, worth trying two.
Updated Oct 13, 2026, every 3 hours. See it live with your plans and costs.
The top 10 for ui & web pages
Front-end pages, components and web apps. Cost per finished task: the tokens at each model's best effort level for this work, plus your time fixing failed attempts ($93 an hour). Success is the share of such tasks a model gets right the first time, on the benchmarks below.
| # | Model | Effort | Success | $/task |
|---|---|---|---|---|
| 1 | Claude Opus 5.5 Anthropic | medium | 67.9% measured | $17.55 |
| 2 | Claude Sonnet 5.5 Anthropic | xhigh | 66.7% measured | $18.44 |
| 3 | GPT-6.1 Sol OpenAI | medium | 63.9% measured | $18.98 |
| 4 | GPT-6 Astra OpenAI | low | 66.1% measured | $22.72 |
| 5 | Gemini 4 Argon Google | high | 60.1% measured | $23.33 |
| 6 | GPT-6 Sol OpenAI | high | 59.0% measured | $24.38 |
| 7 | Claude Fable 5.1 Anthropic | high | 63.8% measured | $25.20 |
| 8 | Claude Opus 5 Anthropic | high | 61.7% measured | $26.04 |
| 9 | Muse Spark 1.3 Meta | xhigh | 55.9% measured | $26.84 |
| 10 | MiMo-V2.6-Pro Xiaomi | default | 53.6% measured | $27.15 |
Best within your tool
| Tool | Model | Effort | $/task |
|---|---|---|---|
| Cursor | Claude Opus 5.5 | medium | $17.55 |
| Claude Code | Claude Opus 5.5 | medium | $17.55 |
| Codex | GPT-6.1 Sol | medium | $18.98 |
| Antigravity | Claude Opus 5.5 | medium | $17.55 |
| GitHub Copilot | Claude Opus 5.5 | medium | $17.55 |
If tokens come with your plan
Tokens included in a plan cost you nothing extra, so the pick is the most reliable model in it, at the leanest effort level within our error of the best (the top levels only use up the plan sooner).
| Plan | Model | Effort |
|---|---|---|
| Claude Pro | Claude Opus 5.5 | high |
| Claude Max | Claude Opus 5.5 | high |
| ChatGPT Plus or Business | GPT-6.1 Sol | high |
| ChatGPT Pro | GPT-6.1 Sol | high |
| Google AI Pro | Claude Opus 5.5 | high |
| Google AI Ultra | Claude Opus 5.5 | high |
Measured on
- LMArena Code Arena – WebDev · LMArena. Blind pairwise votes by developers on web apps built by two anonymous models (Bradley-Terry rating). Measured for 37 of the 42 ranked models, results as of Oct 8, 2026.
- Vibe Code Bench · Vals AI. Build a complete web application from a spec, end to end; scored by automated tests of the running app. Measured for 38 of the 42 ranked models, results as of Oct 12, 2026.
How every number is worked out, with the errors of our estimates: the method.